Doppler

The Doppler effect enables ultrasound to be used to detect the motion of blood (Horskin, 2010). Doppler ultrasound has been used in medicine for almost forty years (Boote, 2003). It is used to measure blood velocity by means of Doppler frequency shift of the echoes received from red blood cells and allows the assessment of tumour vascularity (Rubin, 1994). The concept of Doppler is that malignant neoplasms have active blood vessel creation (angiogenesis) compared to normal or benign neoplasms. Benign lesions tend to form new tumour blood vessels peripherally from pre-existing blood vessels, whereas malignant tumours tend to form new tumour blood vessels centrally, as explained by (Jeong et al., 2000).

There is conflicting evidence as to whether adding colour Doppler imaging to ultrasound screening can reduce the rate of false positive test results or not (Tate et al., 2010). This is due to two issues that arise from colour Doppler ultrasound: first, the assessments are subjective, and second, the assessments depend on the quality of the equipment and the settings used (Timmerman, 2000).

Studies by Guerriero (2001, 1998) focused on the benefits of colour and power Doppler imaging to diagnose ovarian cancer. The earlier study explained that malignancy is suspected by power Doppler when arterial flow is visualized in an echogenic portion of a mass, unlike benign masses that have no similar arterial flow or when flow is seen only at the wall of the mass. It was concluded from that study

48 that power Doppler is helpful when B-mode is indecisive and that it could reduce the number of false positives and thus increase the diagnostic accuracy in atypical cases.

In the later study, it was recommended that at least one of the two Doppler techniques (conventional or power) should be used in conjunction with B-mode imaging as a secondary test (Guerriero et al., 2001, Guerriero et al., 1998).

A later study by (Tempe et al., 2006) examined the usefulness of colour Doppler in the preoperative assessment of ovarian tumours. They concluded that the overall effectiveness in diagnosing the type of lesion is enhanced when adding colour Doppler to ovarian morphology data.

All the studies on colour Doppler imaging revealed a significant overlap in Doppler flow indices between benign and malignant ovarian tumours (Alcázar et al., 2003, Ueland et al., 2003, Van Nagell Jr and Ueland, 1999, Guerriero et al., 1998). Furthermore, Ueland (2003) concluded that the addition of Doppler flow studies did not improve the diagnostic accuracy of the morphologic index.

In 2007, a book was published with a chapter titled: Ultrasound in ovarian carcinoma, in which the author discussed the performance of ultrasound in detecting malignancies based on morphological features and concluded that combining morphological and Doppler ultrasound assessment produces an ideal first imaging test for possible ovarian malignancies. However, the lower specificity of ultrasound requires further imaging evaluation, such as MRI, in patients where ultrasound is inconclusive (Webb, 2007).

49

2.1.2.2.4.1. Doppler indices

In Doppler arterial resistance techniques (Doppler signal analysis), a threshold value is used to characterise the mass. Parameters such as Pulsatility index (PI), resistive index (RI), and peak systolic velocity (PSV) have been used. A recent systematic review of the accuracy of ultrasonography with colour Doppler in ovarian tumours (Medeiros et al., 2009), showed that Doppler can detect malignancy or borderline lesions when RI is below 0.5 with a sensitivity of 87% and specificity of 90%. The authors concluded that Doppler is a useful preoperative test for predicting the diagnosis of pelvic masses and supported an earlier study (Kurjak and Predanic, 1993), which reported that the presence of vessels in the central, septal, or papillary projections, in conjunction with a diffuse vascular arrangement, and RI of less than 0.4, indicated that the mass was expected to be malignant. A possible limitation of this systematic review is the potential bias could be claimed due to all trials included were retrospective and there was a lack of blinding in their assessment.

These results were in agreement with another study (Erdogan et al., 2005) using Doppler ultrasound assessment in the diagnosis of ovarian tumours which showed that detection of Doppler signals in a solid component offered a precise preoperative method to differentiate between benign and malignant ovarian masses.

The major limitation of RI, PI and PSV is that the range of observed measurements in malignant masses overlaps with that observed in benign masses (Kurjak et al., 2003, Myers ER. et al., 2006). Therefore, Valentin et al. (1994) described Doppler as an impractical approach for diagnosing ovarian cancer from a clinical point of view. This argument is supported by another study (Laban et al., 2007), which concluded that RI measurements cannot be used alone for the detection of malignant ovarian tumours.

50 Furthermore, the authors explained that the reason for this is that there is considerable overlap between the RI measurements of benign and malignant ovarian masses. In addition, the overlap in RI range value between benign and malignant masses limits the efficiency of the application of threshold values: thus, cut-off values are not used.

Jeong et al. (2000) explained in their study that a comparison of different studies shows that no standard has been set concerning which Doppler index to use or what cut-off value is most appropriate. However, they found from previous literature that resistive indexes (RI) less than 0.4 and pulsatility indexes (PI) less than 1.0 are generally considered to be suspicious for malignancy. Additionally, the authors disclosed the problems that are associated with Doppler ultrasound, which include operator dependence and lack of standard criteria in distinguishing benign from malignant waveforms. Moreover, certain Doppler indexes can be misleading in premenopausal women due to physiologic alteration in the ovary during the menstrual cycle that cause lowered blood vessel resistance, thereby mimicking malignancy.

2.1.2.2.4.2. Three-dimensional Doppler

A new technique of Doppler ultrasound provides three-dimensional (3D) imaging. Three-dimensional ultrasound was first used in 1989 (Prager et al., 2010). Three- dimensional ultrasound utilizes the real-time capability of ultrasound to build a volume that can be constructed using high-performance work stations. (Hamid et al., 2011, Nelson, 2006).

51 To date, not enough information has been presented to determine whether 3D imaging of the adnexa adds significant information that is not available from conventional two-dimensional (2D) scanning (Benacerraf, 2008).

There are studies reporting that 3D power Doppler ultrasound may be useful for distinguishing benign from malignant ovarian tumours (Kurjak et al., 2003, Alcázar et al., 2005, Testa et al., 2005, Alcázar and Castillo, 2005). For example, Kurjak and colleagues claimed in their study that 3D power Doppler, when combined with the use of 3D sonography, will significantly improve the diagnostic accuracy of detecting stage 1 ovarian cancer, and supported their conclusion by citing evidence from a previous study conducted by (Cohen et al., 2001), which reported that 3D power Doppler better defines the morphological and vascular characteristics of ovarian lesions. Another study supported these finding and disclosed that the accuracy of diagnosing suspected ovarian lesions is significantly enhanced when using 3D ultrasound in combination with 3D power Doppler, and that this approach provides better visualization of tumour vascularity and could significantly improve the diagnostic accuracy in preoperative sonographic assessment in suspected ovarian lesions (Laban et al., 2007).

Alcazar studied tumour vascularity using 3D power Doppler in the early and advanced stages of ovarian cancer and found in his preliminary results that vascularization is higher in advanced stage and metastatic ovarian cancer than in the early stages (Alcázar, 2006).

52 In a previous group of studies in 1999, the researcher and his colleagues found that 3D power Doppler imaging can detect structural abnormalities of malignant tumour vessels, such as arteriovenous shunts. Therefore, it improves and facilitates the morphological and functional evaluation of benign as well as malignant pelvic tumours (Kurjak and Kupešić, 1999, Kurjak et al., 2000a, Kurjak et al., 2000b). Later, the same group of researchers demonstrated the ability of 3D Doppler ultrasound to perform as a secondary test in screening for ovarian cancer and described it as a novel approach for early detection of ovarian cancer (Kurjak et al., 2005).

Other studies suggested the need for further research (Alcázar, 2006, Rieck et al., 2006) and that it should be used as an adjunct to morphologic assessment (Wilson et al., 2006). Although Fishman et. al. (2001) stated that the clinical value of 3D ultrasound is promising for early detection of ovarian carcinoma and need to be investigated deeply, Dai et al. (2008) concluded that it did not improve the diagnostic accuracy for the prediction of malignancy in adnexal masses. Moreover, they further highlighted that 2D transvaginal sonography may still remain an important modality for the prediction of adnexal malignancy. In a preliminary study on 3D analysis of the vascularization of solid masses in the adnexal area, the authors justified the purpose of the study by citing the inaccuracy of 2D B-mode and colour/power Doppler to differentiate between benign and malignant tumours. They concluded that 3D quantitative analysis did not significantly improve the accuracy appreciated by 2D Doppler imaging (Testa et al., 2005).

In addition, Jokubkiene et al. (2007) found that objective quantification of colour signals of the tumour using 3D ultrasound did not appear to add more to B-mode

53 imaging when compared to subjective quantification using 2D power Doppler ultrasound. In a recent review, Alcázar and Jurado (2011) determined that additional studies are necessary to establish the role of 3D ultrasound in clinical practice in gynaecological oncology.